Recent studies have demonstrated that a number of mammalian genes are induced at the transcriptional level as a response to ionizing radiation (IR). Some of these genes are so-called early-response genes encoding pleiotropic transcription factors; other genes may modulate growth, cell cycle progression, deoxyribonucleic acid (DNA) repair and mutagenesis, and ultimately cell survival; and some genes may not be important at all for the recovery after IR. One way to link the induction of these genes with more classical radiobiological endpoints is to-identify IR-induced genes by subtractive complementary deoxyribonucleic acid (cDNA) cloning, then introduce these cDNAs in an antisense (AS) orientation into cells by means of a highly infectious, replication-defective (nonlytic), adenovirus (Ad). The AS ribonucleic acid (RNA) produced by the Ad is expected to inhibit specific cellular gene expression. Since Ad can infect virtually all cells in culture and is transiently maintained in the cells for at least several weeks, one can assess the importance of specifically induced genes in the response to IR. This strategy has several very important and attractive features. For example, all steps can be carried out without prior knowledge of the DNA sequence or even without the complete cDNA. The only requirement would be to clone the cDNA library unidirectionally. The strategy should also be considerably faster than conventional methods, and results generated more informative, since we screen directly for biological function, not identity. To accomplish the above, the applicants will concentrate on the following three objectives: (1) Generation of IR-induced subtraction cDNA libraries at early (3 h) and late (16 h) times after IR from quiescent normal human primary fibroblasts (NF) irradiated with low (2 Gy) and high (20 Gy) doses of 137Cs. (2) Generation of AS Ad by transferring IR-induced cDNAs to Ad by in vivo recombination of cDNA and Ad in human 293 cells. (3) Characterization of specific AS Ad by cell survival (MTT assay, colony forming ability), cell proliferation (3H-thymidine incorporation), cell cycle (flow cytometry), and DNA break/resealing (filter elution) analyses of infected, irradiated, NF.DNA sequencing of positive AS cDNAs and possible gene identification by computer searches. The applicants state that the novel AS Ad approach described will, in a most effective way, help identify those IR-induced genes which are important for cell survival and other cellular processes involved in the recovery from IR. They also-think this technology may be of general use, and not restricted to specific experimental problems in radiobiology.